Process Optimization in Autologous Stem Cell–Derived Islet Generation to Address Key Limitations for Scalability, Safety, and Efficacy

Introduction and Objective: Key challenges in SC-islet scale-up manufacturing includes poor oversight and flawed design-of-experiment (DOE) in quality control (QC) checkpoints, leading to unsafe and failed SC-islet products. Optimal QC parameters for reliable cell production is necessary to ensure clinically safe product. Methods: Patient iPSC-derived islets were produced using vertical-wheel® (VW) bioreactors. QC parameters for stage-specific markers were optimized using flow cytometry, dynamic glucose-stimulated insulin secretion (GSIS) and scRNAseq transcriptomics. Results: In this study, we investigated and addressed key QC challenges for GMP-compliant SC-islet process development. A single 0.5L VW bioreactor vessel achieved a 12-fold increase, yielding a median islet equivalent count of 183,000 (IQR: 255,500-134,300; n=5) and a median cluster size of 159.3 µm (IQR: 91.34-228.2 µm; n=193). We produced 183±14 x 10^6 islet cells from an initial 503±31 x 10^6 iPSCs. FACS-based real-time DOE optimizations at stages -S1, S4, and S6 provided clear differentiation between positive, negative, and false positive populations. FACS-based stage-specific QC revealed a mean 96±5% Sox17+FoxA2+cells at S1-definitive endoderm, 92±1.6% PDX1+NKX6.1+cells at S4-pancreatic progenitor, and 84±1.3% Cpep+NKX6.1+ISL+ at S6- SC-islets that declines to 46.9+1.9% cells following S7 maturation. Additionally, flow-based optimization confirmed a significant reduction in off-target populations (0.6±0.1% OCT4+, 0.5±0.1% Ki67+, 0.5±0.1% CDX2+ cells). Finally, we validated these QC parameters for SC-islet function using dynamic GSIS with over 2-fold insulin secretion and confirmed for β cell identity, functional maturation, and off-target genes using scRNAseq transcriptomics. Conclusion: This study thus provides a comprehensive overview of current limitations, challenges, and solutions for SC-islet manufacturing QC pipeline, offering valuable guidance and considerations. Disclosure: N. Dadheech: None. Z. Czarnecka: None. H. Razavy: None. R.L. Pawlick: None. A. Shapiro: Consultant; Vertex Pharmaceuticals Incorporated, Aspect Biosystems Inc., Betalin Ltd.